> 2.) Subsequent AJAX calls are using a relative path and I don't want
> to hardcode absolute paths in the JS, hence they won't work and the
> application fails since it relies on them.
How paths are represented in URLs does not depend on the framework, but
on the way you build your URLs in the app. I also usually never use any
> I was forgot to save the IRC buffer today so I didn't catch the patch
> but I remembered the zeroing out of Http1, anyway I tried simply doing
> this is in the entrypoint:
> (de start ()
> (off *JS)
> (zero *Http1)
> (if (getUsr) (desktop) (signin)))
> Note the (zero *Http1). That approach seems to be working, I can
In IRC I wrote
(patch _htHead '(format (car @H)) 0) after loading "lib/http.l"
Changing '*Http1' explicitly doesn't help, because it is set by
'_htHead' on each HTTP transaction, depending on the client's HTTP
> One very strange thing though, the HTML output looks like this when
> the form first loads:
> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
> <html xmlns=3D"http://www.w3.org/1999/xhtml" xml:lang=3D"en" lang=3D"en">
> Note the initial 408 and the final 0.
This is because '_htHead' also sets chunked transfer mode.